128
The mutational origin of hybrid sterility genes in rice was demonstrated in irradiation
experiments (Wan and Ikehashi 1996a ). Variety Miyukimochi, which is an irradiated
mutant from Toyonishiki, has two hybrid sterility genes, S5 j^ and S7 j^ , while Toyonishiki
carries a neutral allele, S7 n^ ; therefore S7 n^ must have been mutated into S7 j^ by irradiation.
A second case is that of the experimental line 02428 that has the wide compatibility
allele S5 n^. This line is derived from a progeny population of a hybrid whose parents both
have S5 j^ ; therefore, S5 j^ must have mutated into S5 n^ in both parents as a result of irradia-
tion. The second case is similar to the knock ed- out Gc gene of chromosome 4S in wheat
(Sect. 5.2.2 ), but the fi rst case, i.e., the cre ation of a Gc gene, is not known in wheat.
5.3.3 Evolutionary Implication of the Gametocidal System
Hybrid sterility prevents the movement of genes from one population to the other
within a species, which keeps both populations distinct and eventually leads to spe-
ciation. Suppose two Gc genes of different type, which do not compensate for each
other, are in different populations; hybrids between the two populations will suffer
from sterility due to the gametocidal action. In case the two different Gc genes are
on nonhomologous chromosomes, one-fourth of the gametes produced by the
hybrid become fertile because the nonhomologous chromosomes segregate at ran-
dom in meiosis (cf. Fig. 5.2 ). On the other hand, if the Gc genes are on homologous
chromosomes, all gametes of the hybrid become sterile, because the homologous
chromosomes pair and segregate from each other in meiosis I, and therefore no
gametes will possess both Gc genes. Thus, sexual isola tion would be established in
a species between two populations that easily cross-fertilized.
The formation of new Gc genes or the alteration of existing Gc genes by mutation
is most probable as reported in rice (Wan and Ikehashi 1996a ) and wheat (Friebe
et al. 2003 ). There are various hybrid sterility gene loci in O. glaberrima (Sano
1990 ) and O. sativa (Wan and Ikehashi 1996b ). Gc chromosomes of various
homoeologous groups have been introduced into wheat from different Aegilops spe-
cies, including those fro m the S genome of Ae. sharonensis , which involve
homoeologous groups 2 and 4, and those from the C genome of Ae. cylindrica and
Ae. triuncialis , which are in homoeologous groups 2 and 3, respectively (Endo 1990 ,
2007 ). All the mentioned Gc chromosomes/genes have presumably been involved in
the sexual isolation and speciation of wheat and rice. A suggestive example of sexual
isolation within a species is seen in hybrids between allopatric accessions of Ae.
caudata , which have normal meiotic chromosome pairing, but produce completely
sterile pollen (Ohta 1992 ). This sterility might be explained as the result of the occur-
rence of two different alleles at the Gc loci on homologous chromosomes of the
allopatric accessions. If so, since the Gc alleles segregate during meiosis I into sepa-
rate daughter cells, none of the microgametophytes produced by the hybrids will
receive bot h Gc alleles and will be thus able to develop into fertile pollen.
The presence of incomplete Gc action suggests that the Gc system is involved in the
karyotype evolution of the genus Aegilops. Incomplete Gc action induces chromosomal
T.R. Endo